Testing notes

Nothing is reviewed here on the strength of a press release. Products are used, on faces, for six weeks, and photographed the same way every week. Here is exactly how that is done, including the parts that make our results weaker than they might look.

How we test

We use a split-face design: each tester applies one product to one side of the face and a different product to the other, for six weeks, never swapping sides. Two testers run four products in a single six-week window.

This is a standard dermatology method and it has one consequence you should understand. In a split-face test, each product's result is relative to the product it was paired against, not to a bare face. So every review here names the product that was on the other side. A serum that "won" against a strong competitor and one that won against a weak one are not the same finding, and you should not have to guess which you are reading.

Two testers means two skin types, and both are reported. Two faces are not a population, and we will not write as though they were.

The photographs

Same camera, same room, same light, same distance, same angle — on day zero, every week, and on day forty-two. Never retouched, never filtered, never a flattering angle, and never compared against a photograph taken under different light from its own baseline. If the imagery does not meet that standard, the review does not run.

What we report even when it is unflattering

Stinging, redness, breakouts and the purge phase are logged weekly and published. Copper peptides in particular have a well-known rough patch in the early weeks; reporting it is more useful than pretending everyone sails through.

We also publish the boring, checkable numbers: price per millilitre, the batch we bought, where and when we bought it, and where the active ingredient sits in the ingredient list. Ingredient order is concentration order, which makes it one of the few honest signals available in a category where almost no brand publishes its actual percentages.

What voids a review

Fewer than six weeks. Photographs that cannot be compared to their own baseline. Any retouching. A product given to us free by a brand without that being said on the page. A verdict written before day forty-two. If any of those happen, the review does not publish — we say what we established and what we could not.

Where cosmetics end

Where a trend crosses from cosmetics into research-only compounds, we report the research context and stop. This is a beauty publication, not an instruction manual, and we do not publish doses or protocols for anything injected.